In a Third Generation Partnership Project (3GPP) system, policy and charging functions are implemented by a policy and charging control (PCC) architecture.
The PCC architecture mainly enforces policy and charging control. As shown in FIG. 1, a PCC architecture in the prior art includes an application function (AF), a policy and charging rules function (PCRF), a subscription profile repository (SPR), a policy and charging enforcement function (PCEF), and an online charging system (OCS) or an offline charging system (OFCS).
The AF is adapted to provide access points for service applications and is connected to the PCRF through an Rx interface. Dynamic policy control needs to be performed on the network resources used by these service applications. During parameter negotiation on the service plane, the AF transfers relevant service information to the PCRF. If the service information is consistent with the policy rules defined on the PCRF by operators, the PCRF accepts the service parameters. Otherwise, the PCRF refuses the service parameters and may also carry acceptable service parameters in a response message. Then the AF may return these parameters to users.
The PCRF is adapted to generate policies and charging rules and is connected to the SPR through an Sp interface. As the core of the PCC system, the PCRF provides service data flow based network control rules, including data flow detection, gating, quality of service (QoS) control, and flow-based charging control. The PCRF sends the generated polices and charging rules to the PCEF through a Gx interface and the PCEF charges for service flows accordingly. The PCRF needs to generate policies and charging rules according to the relevant service information obtained from the AF, relevant user subscription data obtained from the SPR, and relevant bearer network information obtained from the PCEF. In addition, the PCRF needs to ensure that these rules are consistent with the user subscription data and to deliver trigger events to the PCEF so that the PCEF may actively request PCC rules from the PCRF when these trigger events occur.
The PCEF is adapted to: enforce the policies and charging rules generated by the PCRF on the bearer plane, receive trigger events delivered by the PCRF so as to actively request PCC rules from the PCRF when these trigger events occur, detect service data flows according to the traffic filter in the rules sent from the PCRF, control these service data flows according to the polices and charging rules generated by the PCRF, and charge for the service data flows online or offline. In online charging mode, the PCEF is connected to the OCS through a Gy interface and works with the OCS to complete credit management. The OCS includes customized applications for mobile network enhanced logic service control point (CAMEL SCP) and a service data flow based credit control. In offline charging mode, the PCEF is connected to the OFCS through a Gz interface and exchanges relevant charging information with the OFCS. In general, the PCEF is located on a gateway (GW) in a network.
The SPR is adapted to store PCC-related user subscription data, including service information that can be used by users, QoS information that can be used by user services, charging-related user subscription data, and group types of users. The PCRF reads the information stored in the SPR through the Sp interface and performs policy control and charging based on user subscription data.
In the PCC architecture, IP-CAN session modification may be initiated by the PCEF. For example, the PCEF may initiate an IP-CAN session modification process when detecting an internal event of IP-CAN session modification caused by an operator's configuration or when detecting a bearer change during data transmission. IP-CAN session modification may also be initiated by the PCRF. For example, the PCRF may initiate an IP-CAN session modification process when the service information at the application layer on the AF changes, when the user subscription data stored in the SPR changes, or when an internal event occurs.
To ensure the competitiveness of the 3GPP system in the coming ten or more years, the 3GPP organization internally proposes system architecture evolution (SAE). In an SAE system, a new PPC architecture including multiple S7 or PCEF interface entities is adopted. In this PCC architecture, multiple S7 or PCEF interface entities are connected to the same PCRF at the same time for any IP-CAN session, which is beyond the capabilities of the PCC architecture in the prior art.
The new PCC architecture includes a roaming PCC architecture and a non-roaming PCC architecture.
FIG. 2 shows a non-roaming PCC architecture in the prior art. The non-roaming PCC architecture includes an AF in a packet data network (PDN), a home PCRF (h-PCRF), a home OCS (h-OCS), a first PCEF (PCEF a), and a second PCEF, PCEF b. The PCEF b is connected to the PDN, h-PCRF, h-OCS, and PCEF a through an SGi interface, an S7b interface, a Gyb interface, and a client mobility IP/proxy mobility IP (CMIP/PMIP) interface respectively. The PCEF a is connected to the h-PCRF and the PCEF b through an S7a interface and a CMIP/PMIP interface respectively. The h-PCRF is connected to the AF through an RX+ interface.
As shown in FIG. 2, the PCEF in the new PCC architecture is divided into two parts: PCEF a and PCEF b. The PCEF a may be configured on an IP access GW such as a serving GW, a PDN GW, or a core network (CN) entity. The PCEF b may be configured on a PDN GW or a CN entity. The bearer concept is not defined on the CMIP/PMIP interface between the PCEF a and the PCEF b. Thus, the bearer-related functions such as bearer binding are configured on the PCEF a. In addition, the SGi interface between the PCEF b and the PDN is a data transmission interface between the CN and the PDN. Thus, the charging and gating functions are configured on the PCEF b.
FIG. 3 shows a structure of a first roaming PCC architecture in the prior art. Different from the non-roaming PCC architecture shown in FIG. 2, the roaming PCC architecture further includes a visited PCRF (v-PCRF) and a visited OCS (v-OCS). As shown in FIG. 3, the PCEF b is connected to the IP Access Network (such as a PDN), h-PCRF, h-OCS, and PCEF a through an SGi interface, an S7b interface, a Gyb interface, and a CMIP/PMIP interface respectively. The PCEF a is connected to the v-PCRF and PCEF b through an S7a interface and a CMIP/PMIP interface respectively. The h-PCRF is connected to the AF and v-PCRF through an RX+ interface and an S9 interface respectively.
In the roaming PCC architecture, the PCEF b may be further divided into a PCEF b1 and a PCEF b2.
FIG. 4 shows a structure of a second roaming PCC architecture in the prior art. Different from the PCEF b in the first roaming PCC architecture shown in FIG. 3, the PCEF b in the second roaming PCC architecture further includes first and second PCEF sub-elements (PCEF b1 and PCEF b2). As shown in FIG. 4, the PCEF b2 is connected to the IP Access Network (such as a PDN), h-PCRF, h-OCS, and PCEF b1 through an SGi interface, an S7b2 interface, a Gyb2 interface, and a CMIP/PMIP interface respectively. The PCEF a is connected to the v-PCRF and PCEF b1 through an S7a interface and a CMIP/PMIP interface respectively. The h-PCRF is connected to the AF and v-PCRF through an RX+ interface and an S9 interface respectively. The PCEF b1 is connected to the PCEF b2 and PCEF a through CMIP/PMIP interfaces respectively. In actual applications, the PCEF b1 may be also connected to the v-PCRF through an S7b1 interface.
The inventor of the present invention discovers that the PCEF in the PCC architecture of the SAE system in the prior art is divided into two parts. That is, the new PCC architecture includes two PCEFs and the PCEF that may perceive a bearer event does not support service data flow based charging. In this case, in online charging mode, it is impossible to ensure that all the PCC rules received by the PCEF that may perceive bearer events pass credit authorization.